Relationship between Dislocation Density and Antibacterial Activity of Cryo-Rolled and Cold-Rolled Copper

In the present work, cold rolling and cryo-rolling were performed on 99% commercially pure copper substrates. Both cold and cryo-rolling processes caused severe plastic deformation that led to an increase in dislocation density by 14× and 28× respectively, as compared to the pristine material. Increases in average tensile strengths, by 75% (488 MPa) and 150% (698 MPa), were observed in the two rolled materials as the result of the enhancement in dislocation density. In addition to strength, enhanced antibacterial property of cryo-rolled copper was observed in comparison to cold rolled and pristine copper. Initial adhesion and subsequent proliferation of bio-film forming Gram-positive bacteria Staphylococcus aureus was reduced by 66% and 100% respectively for cryo-rolled copper. Approximately 55% protein leakage, as well as ethidium bromide (EtBr) uptake, were observed confirming rupture of cell membrane of S. aureus. Inductively coupled plasma-mass spectroscopy reveals higher leaching of elemental copper in nutrient broth media from the cryo-rolled copper. Detailed investigations showed that increased dislocation led to leaching of copper ions that caused damage to the bacterial cell wall and consequently killing of bacterial cells. Cryo-rolling enhanced both strength, as well as antibacterial activity, due to the presence of dislocations.


Introduction
Infection by bacteria via domestic contact is a concern in public areas. Commonly used materials, such as stainless steel, are ineffective in controlling the proliferation of bacteria. In addition to antibacterial activity, the metals used in hospital environments should be non-cytotoxic to human cells. Copper possesses the distinct advantages of being antibacterial and having a non-cytotoxic property [1]. Copper kills bacteria by a mechanism termed as "contact killing", wherein bacterial cells are damaged upon contact with the surface of copper. Release of copper ions from the surface Experimental observations and correlations with metallurgical characterization in cold rolled and cryo-rolled copper were used to explain the phenomenon.

Material Processing
Six-millimeter thick, 99.9% pure copper substrates were subjected to both cold rolling and cryo-rolling in a conventional four high rolling mill (Make: Vaid Engineering Industries, New Delhi, India). The working roller diameter was 142 mm, while the angular velocity of the roller was approximately π rad/s (32 rpm). Commercially pure (99.9%) copper substrates with dimensions of 150 mm × 100 mm × 6 mm (thickness) were used as the raw material (initial).
One set of copper substrates was cold rolled to 1 mm at room temperature with a 5% reduction per pass. The second set of copper substrates was rolled after immersion in a liquid nitrogen bath (maintained at 77 K) for 15 min. The experimental setup for cryo-rolling and cold rolling processes are shown in Figure 1. Due to cryogenic temperatures, only a 4.5% reduction per pass was set. The substrates were rolled immediately. Subsequent cryo-rolling passes were performed after intermittent immersion of the copper substrates in liquid nitrogen for five minutes, as shown in Figure 1. Strength, hardness and wettability properties were evaluated by mechanical tests, while the microstructures were studied by XRD and TEM.

Material Processing
Six-millimeter thick, 99.9% pure copper substrates were subjected to both cold rolling and cryorolling in a conventional four high rolling mill (Make: Vaid Engineering Industries, New Delhi, India). The working roller diameter was 142 mm, while the angular velocity of the roller was approximately π rad/s (32 rpm). Commercially pure (99.9%) copper substrates with dimensions of 150 mm × 100 mm × 6 mm (thickness) were used as the raw material (initial).
One set of copper substrates was cold rolled to 1 mm at room temperature with a 5% reduction per pass. The second set of copper substrates was rolled after immersion in a liquid nitrogen bath (maintained at 77 K) for 15 min. The experimental setup for cryo-rolling and cold rolling processes are shown in Figure 1. Due to cryogenic temperatures, only a 4.5% reduction per pass was set. The substrates were rolled immediately. Subsequent cryo-rolling passes were performed after intermittent immersion of the copper substrates in liquid nitrogen for five minutes, as shown in Figure 1. Strength, hardness and wettability properties were evaluated by mechanical tests, while the microstructures were studied by XRD and TEM.

Physical and Mechanical Characterization
X-ray diffraction studies (Ultima-IV, Rigaku, TX, USA) at a 1.54 Angstrom wavelength using Cu Kα radiation with a scan rate of 4° per minute, in bulk (for 2θ range of 10 to 80°), were performed on 10 mm × 10 mm copper coupons after careful metallographic polishing. The diffraction data were analyzed using PCPDF software and JCPDS database (JCPDS Ref No.: 04-836, 03-105, 78-2076). Contact angle measurements were performed using a DataPhysics OCA 15EC goniometer (DataPhysics Instruments GmbH, Germany). Two-microliter droplets of deionized (DI) water were dispensed on the surface of the coupons. After one minute of stabilization, the contact angles were measured. The micro hardness of the substrates was measured using a Vickers hardness tester (Model: RVM 50PC, Rockwell Testing Aids, India). Uniaxial tensile tests of cold-rolled and cryorolled samples were performed in an Universal Testing Machine (UTM) (Model: 5582, Capacity: 50 kN, Instron, USA) The specimens for tensile testing were profile cut using wire cut electric discharge machining in accordance with the ASTM-E8 standard. For TEM investigations, cryo-rolled and cold rolled copper were gently polished to obtain a foil measuring 80 µm. A blank of 3 mm in diameter

Physical and Mechanical Characterization
X-ray diffraction studies (Ultima-IV, Rigaku, TX, USA) at a 1.54 Angstrom wavelength using Cu Kα radiation with a scan rate of 4 • per minute, in bulk (for 2θ range of 10 to 80 • ), were performed on 10 mm × 10 mm copper coupons after careful metallographic polishing. The diffraction data were analyzed using PCPDF software and JCPDS database (JCPDS Ref No.: 04-836, 03-105, 78-2076). Contact angle measurements were performed using a DataPhysics OCA 15EC goniometer (DataPhysics Instruments GmbH, Germany). Two-microliter droplets of deionized (DI) water were dispensed on the surface of the coupons. After one minute of stabilization, the contact angles were measured. The micro hardness of the substrates was measured using a Vickers hardness tester (Model: RVM 50PC, Rockwell Testing Aids, India). Uniaxial tensile tests of cold-rolled and cryo-rolled samples were performed in an Universal Testing Machine (UTM) (Model: 5582, Capacity: 50 kN, Instron, USA) The specimens for tensile testing were profile cut using wire cut electric discharge machining in accordance with the ASTM-E8 standard. For TEM investigations, cryo-rolled and cold rolled copper were gently polished to obtain a foil measuring 80 µm. A blank of 3 mm in diameter was punched from the foil. The samples were further thinned by ion beam milling (PIPS II, Gaton Inc., Pleasanton, CA, USA) to obtain a transparent region for transmission of the beam. For this purpose, an ion milling machine operating at an accelerating voltage of 5 kV was used for 1.5 to 3 h. Finally, TEM analysis was carried out on a CM12 machine (Philips, Amsterdam, Netherlands) operating at an accelerating voltage of 120 kV.

Adhesion and Proliferation of S. Aureus Cells on Copper Coupons
Staphylococci are prime bacteria causing hospital-acquired infections and are well known to form biofilms on any metal surface that they come into contact with. In general, S. aureus is the most common bacterium responsible for surgical site infections in cardiac and orthopedic surgeries [20,21]. S. aureus expresses poly-β(1-6)-N-acetylglucosamine (PNAG) polysaccharide, a major constituent present in the biofilms of several Gram-positive and Gram-negative organisms [22]. Hence, S. aureus was used as a model bacterium in this study. Coupons, 5 mm × 10 mm in size, with respective thicknesses of the substrate were cleaned by bath sonication for 15 min, each, in acetone followed by deionized water. The coupons were air dried and sterilized by autoclaving at 121 • C for 40 min. S. aureus cell lines (S. aureus, subsp. aureus, ATCC 25923) were maintained on Muller Hinton agar (Hi-Media, India) under standard aerobic conditions at 37 • C. All bacterial experiments were performed under a biosafety level-2 hood.

Bacterial Adhesion and Progression Assay
Sterilized pristine, cold rolled, and cryo-rolled copper coupons were placed in 24-well culture plates in triplicates. One-hundred-microliter inoculums of S. aureus suspensions of 105 CFU/mL concentration were placed on the surface of the copper coupons and incubated for 2 h at room temperature (~25 • C). After the incubation period, non-adherent S. aureus were gently washed twice using 1X phosphate buffer saline (PBS, Sigma Aldrich, USA) at pH 6.8. The copper coupons were placed into fresh 24 well plates and 1 mL of nutrient broth (Hi-Media, India) was added into the wells. Three sets of coupons were further incubated at 37 • C, each for time intervals of 0, 24 and 72 h.

Bacterial Adhesion and Progression Assay
After the incubation period, copper coupons were removed from the respective wells and gently washed with sterile 1X PBS to remove non-adherent cells. The copper coupons were placed in glass tubes containing 1 mL of sterile DI water. The tubes were vortexed for 2 min and sonication for 15 min to obtain a homogenous bacterial cell suspension from the coupons. The obtained suspensions were serially diluted to seven-fold dilutions in 1X PBS (pH 6.8). A volume of 100 µL of the diluted suspension was spread plated on Muller-Hinton agar plates in triplicates. The agar plates were incubated at 37 • C for 24 h. Colony-forming units (CFUs) were enumerated and the results were expressed in CFU/mL. The bacterial reduction (BR) rate was calculated as the ratio of the difference in colonies on the test coupons (cold-rolled and cryo-rolled) to the number of colonies on pristine coupons at different time intervals.
The disruptive effect of copper on S. aureus was assessed by ethidium bromide (EtBr) uptake assay. S. aureus treated with 1% Triton X-100 was used as a positive control whereas cells with EtBr not exposed to copper substrates served as negative controls. Ten microliters of each suspension were examined under a fluorescence microscope. Inductively coupled plasma mass spectrometry (ICP-MS) was used to estimate the leaching of elemental copper from pristine, cold rolled and cryo-rolled copper coupons in the nutrient broth. Detailed information on the toxicity, other cellular studies and leaching of copper ions are given in the supplementary section.

Results and Discussion
The average micro-Vickers hardness of pristine, cold rolled, and cryo-rolled were 70 HV, 130 HV, and 180 HV respectively. The tensile strength of the pristine, cold rolled and cryo-rolled were 279 MPa, 488 MPa and 698 MPa respectively (as reported in the Supplementary Information, Figure S2). Cryo-rolling was performed in 39 passes while cold rolling was performed at 35 passes. It has been established that the slower rate of strain hardening rate also contributes to the increase in strength [23]. The crystalline size (d) is computed using Williamson-hall method [24] and is used to calculate dislocation density [25,26]. The peaks of Cu (111) and (220) broadened after cryo-rolling due to the combined effects of reduced crystallite size (d), increased residual strain (ε) and increased dislocation density (ρ) are reported in Table 1. The crystalline size (d) was calculated using the Williamson-Hall equation [27], as per Equation (1), where ε is micro-strain, θ is diffraction angle and B is peak broadening width. These parameters (ε, θ, and B) were determined by examining the XRD peak broadening using well-known XRD analysis software (PANalytical X'pert Highscore). To improve the accuracy of XRD analysis, careful filtration of unwanted instrumental broadening was done by analyzing the XRD peaks of a standard specimen using the deformed copper samples (cold-rolled and cryo-rolled). The dislocation density (ρ) was calculated using Equation (2) [25,28], where b is the magnitude of Burgers vector. The increase in the Vickers hardness and the tensile strength is a result of smaller crystalline sizes and an increase in dislocation density (Table 1). After the initial exposure, cryo-rolled coupons exhibited reduced adhesion compared to cold rolled and pristine coupons, although the contact angle (usually exhibits inverse correlation with adhesion) measured was lower (as shown in Figure S1 in the Supplementary Information). On further exposure for 24 and 72 h, cryo-rolled coupons exhibited antibacterial activity. The retrieved cells from the coupons were cultured on MacConkey ® agar plates for 24 h (Figure 2). However, both cold rolled and pristine copper did not inhibit adhesion and progression of S. aureus. The adhesion was reduced by 7.73% and 67.11%, respectively, in the cold-rolled and cryo-rolled samples during initial exposure. After the 24-h exposure, cryo-rolled coupons demonstrated a 100% reduction whereas cold-rolled coupons exhibited a mere 8.02% reduction. After 24 h growth, it was observed that S. aureus established a higher CFU on pristine than on cold-rolled or cryo-rolled copper coupons, as shown in Figure 3. The number of bacterial cells per colony decreased drastically for cryo-rolled copper, indicating significant impairing of bacterial cell growth and colony formation (Figure 4). Though S. aureus cells were observed on cryo-rolled copper coupons at 0 h, they were not viable, as seen in Figure 2.   S. aureus cells exposed to cryo-rolled coupons exhibited increased staining 4-fold of EtBr as compared to pristine and cold-rolled copper ( Figure S4 in Supplementary Information). The mechanism responsible for this improvement is not been established thus far. On the other hand, studies on the biocompatibility of ultrafine-grained materials in recent years [29] suggest a better adhesion of cells, given the material's wettability (from contact angle measurements) [30]. This   S. aureus cells exposed to cryo-rolled coupons exhibited increased staining 4-fold of EtBr as compared to pristine and cold-rolled copper ( Figure S4 in Supplementary Information). The mechanism responsible for this improvement is not been established thus far. On the other hand, studies on the biocompatibility of ultrafine-grained materials in recent years [29] suggest a better adhesion of cells, given the material's wettability (from contact angle measurements) [30]. This    S. aureus cells exposed to cryo-rolled coupons exhibited increased staining 4-fold of EtBr as compared to pristine and cold-rolled copper ( Figure S4 in Supplementary Information). The mechanism responsible for this improvement is not been established thus far. On the other hand, studies on the biocompatibility of ultrafine-grained materials in recent years [29] suggest a better adhesion of cells, given the material's wettability (from contact angle measurements) [30]. This  S. aureus cells exposed to cryo-rolled coupons exhibited increased staining 4-fold of EtBr as compared to pristine and cold-rolled copper ( Figure S4 in Supplementary Information). The mechanism responsible for this improvement is not been established thus far. On the other hand, studies on the biocompatibility of ultrafine-grained materials in recent years [29] suggest a better adhesion of cells, given the material's wettability (from contact angle measurements) [30]. This conflicts with our present results. It is to be, however, clarified that the hydrophilicity confirmed via contact angle does not guarantee the chemical affinity between the cell wall and adhering surface.
Contact killing by Cu 1+ /Cu 2+ ions is the widely-accepted mechanism for copper's antibacterial property. Since the materials investigated differ only in their substructures (Table 1), it is pertinent to explore the influence of dislocation density on the ionization of copper. Ionization of copper, when in contact with aqueous medium, leads to the creation of static charge field around the dislocations [31]. Charged dislocations are generally surrounded by a cloud of charged vacancy defects 'q' [32]. The interaction between this charge (q) and dislocation can be correlated with the ease of Cu ion formation. The dislocation-charged cloud interaction energy (E int ) [33,34] is expressed as: where G and ν are elastic constants, ε is the relative permittivity and x and y are the position vectors of the dislocation. The lower the E int , the easier is the formation of ions. Since the elastic constants in E elas are invariant, E int differs only by the electrostatic coulomb interaction energy (E Coul ). The charge, 'q' around the dislocation can be estimated using the Koehler equation [35].
where ζ is the static charge constant value, T is the temperature k is the Boltzmann constant, a is the material constant, 's' is the mean free path of dislocation and ∂z ∂s is the field potential gradient around the dislocation. To understand the ease of ion formation, it is necessary to relate the dislocation potential (z) to the dislocation density, as given by Pödör [36].
where ρ is the dislocation density, m e represents electron effective mass, c is the lattice constant, and β is the inverse screening length. Among the three copper samples, ρ is the only variable in Equation (5), therefore: ∂z ∂s = φρ (6) where φ = z/ρ s. Substituting Equation (6) into Equation (4), using Equation (7), E Coul can be expressed as Since cryo-rolled material has the highest dislocation density (Table 1), the E int of this material will be lower as per Equations (3), (7) and (8). This will result in higher extent of Cu ion formation (Cu 2+ and Cu + ) in cryo-rolled material leading to better antibacterial property. The antibacterial property of the three samples can be directly correlated to the dislocation density. The quantitative results of inductively coupled plasma mass spectrometry (ICP-MS) indicate that elemental copper leached out from pristine, cold rolled and cryo-rolled copper coupons in the amounts of 644.5, 975.9, and 1441.5 ppb/mm 2 respectively, into the nutrient broth (Supplementary Information). The increased leaching of copper from cryo-rolled copper as compared to cold rolled and pristine supports the above discussion. The increased ion density of cryo-rolled copper, confirmed by leaching, has led to higher contact killing S. aureus. However, the cryo-rolled copper proved to be non-toxic to humans in a hemolysis experiment (as shown in Figure S5 in the Supplementary Information).
Both cryo-rolled and cold rolled materials were investigated for the sites of dislocations by transmission electron microscopy (TEM). In cold-rolled and cryo-rolled materials, the micrograph represents several sites with densely packed dislocation tangles/ clusters (Figure 5a,c). In the cryo-rolled material, dislocation tangle/clusters appear to be denser in Figure 5d, as compared to the cold rolled material shown in Figure 5b. In addition to the dense dislocation cluster, cryo-rolled materials also exhibit closely packed nanotwins surrounded by packets of dislocation Figure 5d,e. The reason for such nanotwin in cryo-rolled can be attributed to (i) low stacking fault energy of copper; (ii) cryogenic processing temperature; and (iii) movement of Shockley partials via cross slip. The nanotwins in the form of Shockley partials also contribute to the overall dislocation density of the cryo-rolled material. On the basis of all of the above results, we conclude that the cryo-rolled copper contains denser dislocation clusters and nanotwins (as reported in Figure 5). The higher dislocation density results in enhanced leaching of copper ions due to direct effect of dislocation density on the charges. The enhancement of the leaching of copper ions was verified by the ICP-MS results. The leaching of copper ions helps in the disruption of the cell wall of S. aureus bacteria, the EtBr uptake assay and protein leakage results ( Figure S3 in the Supplementary Information) shows significant evidences of damage of cell-wall of bacteria during interaction of copper surface with adhered bacterial species. Therefore, excessive leaching of copper ions from high strength cryo-rolled samples promotes contact killing of S. aureus and thus results into improved antibacterial property of the cryo-processed copper. Due to enhancement of antibacterial property of high strength cryo-rolled copper, initial adhesion and subsequent proliferation S. aureus were reduced by 66% and 100%, respectively, as shown in Figure 3.

Conclusions
In summary, it can be stated that the cryo-rolling process for strengthening also enhances the antibacterial property of copper. The dislocations in cryo-rolled copper are surrounded by positively-charged electrostatic clouds, that in-turn leads to an increased internal energy. In this first reported work, experimental observations on bacterial adhesion and cell wall disruption indicate that cryo-rolled copper is the most effective antibacterial material as compared to cold rolled and pristine copper. An in-depth investigation of enhanced antibacterial property of the cryo-rolled copper revealed that denser dislocations are induced by the cryo-rolling process which lead to release of excessive copper ions as compared to pristine copper. These copper ions damage the bacterial cell wall, resulting in the contact killing of bacterial cells and therefore the cryo-rolling process improves antibacterial property significantly. The major contribution of the present work is the relationship between dislocation density, leaching of copper ions and the antibacterial property.